Objective： American ginseng is a medicinal plant with large market demands,however,its producing areas are shrinking because of the continuous cropping obstacles in China.Therefore,it is urgent to establish a suitabl...Objective： American ginseng is a medicinal plant with large market demands,however,its producing areas are shrinking because of the continuous cropping obstacles in China.Therefore,it is urgent to establish a suitable model to determine the new producing areas.Here we evaluated and predicted the suitable areas of American ginseng using the maximum entropy model（Max Ent）.Methods： Based on the 37 environmental variables over thirty years from 1970 to 20 0 0 and 226 global distribution points of American ginseng,Max Ent was used to determine the global ecological suitable areas for American ginseng.The Receiver Operating Curve（ROC）was used to evaluate the model prediction accuracy.Meanwhile,an innovative ecological variable,the precipitation–temperature ratio,was established to indicate the climate characteristic in the American ginseng suitable areas based on the monthly precipitation and temperature.Results： The potential ecological suitable areas of American ginseng were primarily in Appalachian Mountain in America and Changbai Mountain in China,about in the range of 35 °N–50 °N,60 °W–120 °W and 35 °N–50 °N,110 °E–145 °E,respectively,including the United States,Canada,China,North Korea,South Korea,Russia and Japan.South Korea and Japan were the potential producing regions.The precipitation–temperature ratios were stable at（0.22,0.56）of the vigorous growth period（April–October）in the best suitable areas of American ginseng,serving as characteristic parameters to optimize the prediction model.The model showed that the common soil parameters were pH 4.5–7.2,Base Saturation（BS）above 80%,Cation Exchange Capacity（CEC）10–20 cmol/kg,organic carbon（OC）〈 1.4%,and the soil types were sandy loam or loam.Conclusion： An optimized Max Ent model was established to predict the producing area for American ginseng that needed to be validated by a field test.展开更多
基金supported by National Natural Science Foundation of China (81473304)National Science and Technology Support Program (2015BAI05B01)
文摘Objective： American ginseng is a medicinal plant with large market demands,however,its producing areas are shrinking because of the continuous cropping obstacles in China.Therefore,it is urgent to establish a suitable model to determine the new producing areas.Here we evaluated and predicted the suitable areas of American ginseng using the maximum entropy model（Max Ent）.Methods： Based on the 37 environmental variables over thirty years from 1970 to 20 0 0 and 226 global distribution points of American ginseng,Max Ent was used to determine the global ecological suitable areas for American ginseng.The Receiver Operating Curve（ROC）was used to evaluate the model prediction accuracy.Meanwhile,an innovative ecological variable,the precipitation–temperature ratio,was established to indicate the climate characteristic in the American ginseng suitable areas based on the monthly precipitation and temperature.Results： The potential ecological suitable areas of American ginseng were primarily in Appalachian Mountain in America and Changbai Mountain in China,about in the range of 35 °N–50 °N,60 °W–120 °W and 35 °N–50 °N,110 °E–145 °E,respectively,including the United States,Canada,China,North Korea,South Korea,Russia and Japan.South Korea and Japan were the potential producing regions.The precipitation–temperature ratios were stable at（0.22,0.56）of the vigorous growth period（April–October）in the best suitable areas of American ginseng,serving as characteristic parameters to optimize the prediction model.The model showed that the common soil parameters were pH 4.5–7.2,Base Saturation（BS）above 80%,Cation Exchange Capacity（CEC）10–20 cmol/kg,organic carbon（OC）〈 1.4%,and the soil types were sandy loam or loam.Conclusion： An optimized Max Ent model was established to predict the producing area for American ginseng that needed to be validated by a field test.
